Extreme Learning Machine based Linear Homogeneous Ensemble for Software Fault Prediction

Pravas Ranjan Bal, Sandeep Kumar

2018

Abstract

Many recent studies have experimented the software fault prediction models to predict the number of software faults using statistical and traditional machine learning techniques. However, it is observed that the performance of traditional software fault prediction models vary from dataset to dataset. In addition, the performance of the traditional models degrade for inter release prediction. To address these issues, we have proposed linear homogeneous ensemble methods based on two variations of extreme learning machine, Differentiable Extreme Learning Machine Ensemble (DELME) and Non-differentiable Extreme Learning Machine Ensemble (NELME), to predict the number of software faults. We have used seventeen PROMISE datasets and five eclipse datasets to validate these software fault prediction models. We have performed two types of predictions, within project defect prediction and inter release prediction, to validate our proposed fault prediction model. The experimental result shows consistently better performance across all datasets.

Paper Citation

in Harvard Style

Bal P. and Kumar S. (2018). Extreme Learning Machine based Linear Homogeneous Ensemble for Software Fault Prediction.In Proceedings of the 13th International Conference on Software Technologies - Volume 1: ICSOFT, ISBN 978-989-758-320-9, pages 69-78. DOI: 10.5220/0006839500690078

in Bibtex Style

@conference{icsoft18,
author={Pravas Ranjan Bal and Sandeep Kumar},
title={Extreme Learning Machine based Linear Homogeneous Ensemble for Software Fault Prediction},
booktitle={Proceedings of the 13th International Conference on Software Technologies - Volume 1: ICSOFT,},
year={2018},
pages={69-78},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006839500690078},
isbn={978-989-758-320-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Software Technologies - Volume 1: ICSOFT,
TI - Extreme Learning Machine based Linear Homogeneous Ensemble for Software Fault Prediction
SN - 978-989-758-320-9
AU - Bal P.
AU - Kumar S.
PY - 2018
SP - 69
EP - 78
DO - 10.5220/0006839500690078